GB2434717A

GB2434717A - Formation of networks

Info

Publication number: GB2434717A
Application number: GB0601622A
Authority: GB
Inventors: Georgios Kalogridis; Neil Fanning
Original assignee: Toshiba Research Europe Ltd
Current assignee: Toshiba Europe Ltd
Priority date: 2006-01-26
Filing date: 2006-01-26
Publication date: 2007-08-01
Anticipated expiration: 2026-01-26
Also published as: GB2434717B; GB0601622D0

Abstract

A facility in a MAC layer of a wireless communications network provides two modes of operation of a piconet. In a first mode, the medium is offered for contention by all devices in a substantially conventional fashion. In a second mode, the medium is offered for contention by clusters of devices in turn. In the latter case, the contention period is sub-divided and subdivisions are allocated to respective clusters. All devices in the network remain capable of receiving communication from all other devices, regardless of which mode is in operation.

Description

1 2434717 Wireless Communications Apparat The present invention relates

to the field of wireless communications and in particular relates to the formation of networks and associated communications devices. The invention particularly but not exclusively, relates to the establishment and management of ad hoc networks, and to communications devices associated therewith.

For example, an ad hoc network is defined as a network composed solely of stations within mutual communications range of one another via a wireless medium, e.g. wireless personal area networks (WPAN) which convey information over relatively short distances amongst a relatively small number of participants. It is noted that in comparison to wireless local area networks (WLAN), WPAN arrangements involve little or no infrastructure.

The term "piconet" was originally introduced in relation to Bluetooth specifications but has now more generally been defined in IEEE8O2.15.3 (Wireless MAC and PHY specifications for 1-ugh Rate WPANs) as a collection of one or more logically associated devices that share a single identifier with a common coordinator. A Piconet coordinator is an entity that has device functionality but which also provides coordination and other services (e.g. quality of service [QoS], synchronisation association etc) via the wireless medium for the "one or more" devices within the piconet. An ad hoc wireless network can be considered as a piconet.

Generally in wireless technologies such as WPAN (e.g. Bluetooth or IEEE 802.15.1 or IEEE 802.15.3), WLAN (e.g. 802.1 lx), ultra-wide broadband (UWB) etc, a common radio channel is used which is then shared by all the devices in the network (be they in a coordinated ad hoc network, a piconet or in an uncoordinated neighbouring network).

The OSI, or Open System Interconnection, model defines a networking framework for implementing protocols in seven layers. Control is passed from one layer to the next, starting at the application layer (layer 7) in one station, proceeding to the bottom layer, over the channel to the next station and back up the hierarchy.

The bottom layer, layer 1, is known as the Physical (PHY) layer. This layer conveys the bit stream (which could be an electrical impulse, light or radio signal) through the network at the electrical and mechanical level. It provides the hardware means of sending and receiving data on a carrier, including defining cables, cards and physical aspects. Fast Ethernet, RS232, and ATM are protocols with physical layer components.

Layer 2 is known as the Data Link layer and at this layer, data packets are encoded and decoded into bits. It furnishes transmission protocol knowledge and management and handles errors in the physical layer, flow control and frame synchronization. The data link layer is divided into two sublayers: The Media Access Control (MAC) layer and the Logical Link Control (LLC) layer. The MAC sublayer controls how a computer on the network gains access to the data and permission to transmit it. The LLC layer controls frame synchronization, flow control and error checking.

Layer 3, the Network layer, provides switching and routing technologies, creating logical paths, known as virtual circuits, for transmitting data from node to node.

Routing and forwarding are functions of this layer, as well as addressing, internetworking, error handling, congestion control and packet sequencing.

In the wireless technologies mentioned above (WPAN, WLAN, UWB etc) the PHY layer coordinates sharing of radio resources between different piconets. A wireless device is normally able to set up or join a distinct wireless network and multiple piconets should, ideally, be able to coexist with one another.

The PHY layer shares the wireless medium between piconets by negotiating parameters such as the modulation scheme, the transmission power and the logical channel(s) to occupy. It is the aim of the PHY later that the wireless medium is used efficiently and that interference between piconets is minimised.

The MAC sub-layer is conventionally found on top of the PHY layer and is responsible for managing one piconet's share of the wireless medium. The MAC layer may support direct communications among devices or communications via a master control station.

Additionally, it will be responsible for complex management procedures such as association, disassociation, re-association, hand-over, etc. Its major task, however, is to organise access to the available bandwidth in the most efficient way. This could either mean exploiting as much of the available bandwidth as possible, supporting prioritised traffic and QoS, being fair to all participants, and doing all these with the minimum time latency and bandwidth overhead. The MAC needs also to be adapted to special PHY characteristics and to exploit PHY features that will improve its performance.

Some wireless networks, e.g. multi-hop ad hoc networks, also require routing protocols (which are handled by the NETWORK layer) to optimise piconet formation and traffic routes.

The traditional set-up for wireless networks can therefore be regarded as two level sharing between the PHY layer (for sharing resources between co-existing wireless networks) and the MAC layer (for sharing resources among devices within a specific wireless network). In this traditional methodology the PHY channel is engineered to cover as many wireless devices as possible and a sophisticated MAC protocol is designed to take advantage of the PHY channel.

The above approach makes no allowance for relative levels of communication between devices. All devices are assumed to require the facility to communicate with all others in the piconet, without limitation. This can lead to inefficiency and can give rise to contention issues in the piconet in operation. However, simply breaking the piconet into a plurality of smaller piconets reduces the flexibility of the network to allow, when occasion arises, the communication of one device with another, regardless of the rarity of such communication.

It is therefore an object of the present invention to provide methods and apparatus which substantially mitigate or overcome the above mentioned problems.

An aspect of the invention provides a wireless MAC protocol which provides for contention for medium access by means of clustering devices within the piconet. The clustering in this aspect of the invention is not intended to cause the formation of separate piconets, but provides for more effective contention where it is most often required.

Another aspect of the invention provides a communication device wireless network comprising control means for controlling access to a medium in the network, and a plurality of wireless devices assigned to the network, wherein the control means is operable to arrange said devices into two or more clusters operating in the same network.

Another aspect of the invention provides software means operable to configure a computer provided with general purpose communications means to implement the wireless MAC protocol set out above. The software means may be provided by means of a computer program product, which may be supplied by way of a data storage device or borne on a signal for instance downloaded to the computer by a prior established communications channel.

Further aspects of the invention will become apparent from the following description of a specific embodiment of the invention. It will be appreciated that the specific embodiment is described by way of example only, and should not be viewed as limiting the scope of the invention, which is to be considered as defined by the appended claims as read in conjunction (but not limited by) the description and accompanying drawings, in which: Figure 1 is a schematic diagram of a conventional wireless MAC protocol; Figure 2 is a schematic diagram of a networking environment provided to illustrate a specific embodiment of the present invention; Figure 3 is a schematic diagram of a computer communications device provided in the networking environment illustrated in figure 2; Figure 4 illustrates a state transition diagram for use in the wireless MAC protocol for implementation in the networking environment illustrated in figure 2; Figure 5 illustrates allocation of a contention access period (CAP) in accordance with a first mode (mode 0) of operation of the wireless MAC protocol of the specific embodiment; Figure 6 illustrates a CSMA/CA block for access to a transmission opportunity in the CAP illustrated in figure 5; Figure 7 illustrates allocation of a contention access period (CAP) in accordance with a second mode (mode 1) of operation of the wireless MAC protocol of the specific embodiment; Figure 8 illustrates the networking environment illustrated in figure 2 indicating communications links utilised in accordance with operation mode 0; Figure 9 illustrates the networking envornment illustrated in figure 2 indicating communications links utilised in accordance with operation mode 1; and Figure 10 illustrates a MAC manager implemented in one, or more than one, of the communications devices illustrated in the networking envornment in figure 2.

Presently, a conventional wireless MAC protocol is illustrated in Figure 1. This uses a time-divided medium in order to organise access of wireless devices within a piconet.

The medium is time-divided by means of superframes, each of which is commenced with a beacon, following which a contention access period ensures and thereafter a channel time allocation period takes the remainder of the superframe.

The contention access period (CAP, or contention period in IEEE8O2. 11 nomenclature) is used for carrying commands and non-stream data, while the channel time allocation period (CTAP, or contention free period in IEEE8O2. 11 nomenclature) is used for a synchronous and isocbronous data streams. For example, the CAP can be used for random traffic (which may or may not be transmitted in bursts), for example such traffic attributed to HTTP or FTP protocols. The CTAP can thus be used for applications that require a certain quality of service, for example video, music or voice data. A device will be able to gain access to the CAP by contending.

Furthermore, the boundaries between the CAP and the CTAP can be flexible and adjusted in accordance with the current usage of the network. For example, if no non-stream data is being transmitted, then there may be no need for a CTAP and, conversely, if only data streams are being transmitted, a CAP may not be required.

Figure 2 illustrates a specific embodiment of the invention, comprising a networking environment 10 in which are located a plurality of wireless communications devices 20 and a wireless communications controller 22. It will be understood that, whereas the invention is described herein with regard to a centrally controlled medium access layer, a distributed control apparatus could equally be provided.

The networking environment provides facility for ad hoc wireless communication to be established. This is achieved by means of communications control facilities provided on the communications controller 22, as will now be described with reference to Figure 3. Figure 3 illustrates the controller 22 as being implemented by means of a general purpose computer. In this case, communications facilities are provided by means of hardware, which is in turn configured by means of software. More particularly, the controller comprises a processor 30, in communication with the working memory 32 and a bus 34. A mass storage device (which, in this case, is a magnetic storage device, though other such storage devices would suffice) 36 is provided for long term storage of data andlor programs not in immediate use. A medium access controller 38 is connected to an antenna array 40, to provide the controller 22 with access to the wireless communications medium. Further, a broadband modem 42 is provided to allow the communications controller to be connected to hardwired communications networks, such as the internet. This may be used to provide the network established by the wireless communications controller 22 with a portal to the internet.

In conventional maimer, the controller further comprises audiovisual output devices 44 and user operable input devices 46.

Between the working memory 32 and the mass storage device 36, the controller 22 stores executable programs enabling the processor 30 to configure the controller 22 to provide medium access control in accordance with the specific embodiment of the invention as will be described below. As shown in Figure 4, the present example implements two modes of operation within the wireless controller 22, namely modes 0 and 1. A mode selector, implemented in the controller 22, is operable to determine whether it is likely that mode 0 will perform better than mode I. The state diagram illustrated in Figure 4 sets out the responses that the mode selector will make depending on the result of that enquiry. For example, if the controller is already operating in mode 0, then if mode 0 is likely to perform better than mode 1, then the mode selector makes no change to the mode selection. Otherwise, mode 1 is selected. Likewise, in mode 1, if the enquiry shows that it is likely that mode 0 will operate more effectively than mode 1, then the mode selector changes the mode to mode 0. Otherwise, in mode 1, no change in selection is made.

Operation of the wireless MAC controller 22, and thus the piconet established thereby, in mode 0 is entirely conventional. That is, as shown in Figure 5, the CAP is divided into a succession of transmission opportunities or time slots. Access to these time slots is achieved by means of application of a repeat cycle of CSMAICA, with or without RTSICTS (which are entirely optional). As will be appreciated by the skilled person, each block of CSMA/CA consists of inter-frame spaces (IFS) which can have various lengths according to the particular application, contention window space (CW), data and acknowledgement (ACK) frames, and optionally RTS and CTS frames. Figure 6 illustrates a typical CSMAICA block, without RTS/CTS frames.

Operation of the piconet established by the controller 22 with the other wireless communications devices 20 in accordance with mode 0 is conventional and will be understood by the skilled person without further description. The controller 22 will establish and manage communication between devices along the inter-device communications links indicated in figure 8.

Operation of the piconet in accordance with mode 1 will now be described.

In mode 1, the piconet devices form two logical clusters. It will be appreciated by the skilled person that more than two logical clusters could be established, although two clusters are illustrated in Figure 9. Each cluster is associated with a respective sub-CAP, as illustrated in Figure 7. Each sub-CAP can then use a customised CSMAICA (with or without RTS/CTS) mechanism, within its own time slot. The relative lengths of the multiple sub-CAPs will vary and will depend on the traffic needs within each cluster. Relative lengths are managed by the controller 22.

In order to access the CAP, a device contends for access within the sub-CAP period assigned to the cluster to which that device is itself assigned. In accordance with this specific embodiment, a device cannot access the CAP outside that respective sub-CAP period. On the other hand, all devices in the piconet remain capable of reading the whole CAP, so that a device can receive data from any other device, not just those within its own cluster. Therefore, it will be appreciated that the clustering of devices in this way does not constitute splitting of the piconet into separate sub-piconetS. In essence, the clustering provides a facility to ensure that devices can be grouped together in an advantageous fashion to ensure that those devices which require access to the CAP can do so without contending with other devices of another cluster which would otherwise restrict access to the CAP.

Control of this process is carried out by an MAC manager 100, which may be implemented by means of a computer program, for instance retrievable from the mass storage unit 36 of the controller 22. The computer program will be executed by the processor 30 making use of working memory 32. Figure 10 illustrates schematically the arrangement of the MAC manager 100 within the controller 22.

The MAC manager 100 interacts with the MAC protocol layer of the communications stack defined across the piconet. The MAC protocol layer 110 is illustrated schematically in Figure 10 as being hosted on the controller 22, but it will be appreciated that this is merely the instance of the MAC protocol layer on the controller 22 and other instances will reside at the other modes 20 of the network. Further, a network layer 112 is illustrated, which in practice will be of similar arrangement.

The network layer is capable of being monitored by the controller 22 to provide traffic analysis data and the requirements of applications accessing the network layer 112.

This data 120, as indicated in Figure 10, is stored in working memory in the controller.

In use, therefore, the MAC manager 100 determines dynamically which mode, out of modes 0 and 1, should be employed. The method in which this is carried out, in accordance with the specific embodiments of the invention, will now be described.

The probability ir that device i sends non-stream data to device] is determined for all device pairs. This probability is dynamically calculated and updated in real-time by the MAC manager 100 by analysing traffic in the piconet.

Mode 0 will be preferred in the following situations: * 7t has a substantially uniform distribution; or * is unknown and no judgement can be made; or * the overall non-stream traffic is low and the CAP is clearly not close to saturation Mode I will be preferred when both the CAP (in mode 0) is relatively close to saturation and when the following conditions arise: * There exists one group A{al, a2, . . . }, which has the following properties: o A has at least two members ak o ak represents a device that is a node of the piconet o A has at most two devices less that the number of devices that comprise the piconet o for every i that belongs in A, and for every j that doesn't belong in A, 7tij is very low relative to respective probabilities that apply for the devices that are members of A. A practical example of this example scenario demonstrates its use. A WLAN/WPAN piconet is provided, consisting of 6 wireless devices {al, a2, a3, a4, a5, a6}. al is a laptop backing up its whole hard disk to a2, while a2 moves a large amount of data to a3. a4 operates as a gateway to the internet, so all other devices wishing to access the internet do so through a4. al, a2 and a3 only have very little internet usage, while a5 and a6 are rather heavy internet users.

Using this information, the following sub-groups: A{al, a2, a3} and B={a4, a5, a6} are formed. A satisfies all the properties mentioned above, since: * t12, 23 are high; * 7t45, 7t46 are high; * 1t14, 7t24, 7t34 are low; and * The remaining probabilities are, relative to these, very low or zero.

In this case two sub-CAPs will be defined in the CAP, respectively allocated to groups A and B. Each sub-CAP will have fewer active contending devices (3 instead of 6), hence providing the potential for improvement in bandwidth exploitation, allowing for better performance and higher saturation levels.

It will be appreciated that this is but one specific example and other scenarios could be designed for management of the transition between mode 0 and mode 1.

In the present illustrated example, as illustrated in figure 9, two clusters are formulated which operate in accordance with MAC mode 1. The first cluster consists of all devices that are designated as being members of group A. The second cluster consists of all remaining devices in the piconet.

The dotted line 130, indicated in figure 9 between the two clusters, connects two arbitrary inter-cluster devices 20 and represents the low traffic envisaged between these devices. Inter-cluster communication is implemented in a straightforward manner, since clusters co-exist -they are imposed on the piconet rather than causing formation of further piconets. It is emphasised that, in accordance with this specific embodiment, all devices (and, therefore, all clusters of devices) belong to the same piconet. Their PHY (1hysical layers) are synchronized on the same channel. All devices are therefore able to read the two (or more) MAC sub-CAPs, as previously discussed. This procedure is then iteratively applied and further clusters can be formed of any (sub)-group of devices in the piconet. The lengths of the sub-CTA timeslots can be determined by the MAC manager 100 to be a function of the traffic that is calculated to take place within each cluster.

If inter-cluster traffic increases, then queues in nodes in each sub-CAP will grow longer, more collisions will take place, MAC performance will suffer and the MAC manager will eventually determine that MAC mode 0 will probably perform more effectively. In that case, the MAC manager will abandon the clustering and will allow all devices to contend for access of the CAP at any time.

While the invention has been described in the context of application to communications in accordance with the IEEE8O2.15.3 and IEEE8O2.11 families of standards, and may indeed form part of a performance enhancement thereto, it will be appreciated that it is not intended that the invention is limited thereto. Indeed, the invention can be applied to MAC specifications of WPAN, WLAN, Ultra Wideband (UWB) or ad hoc networks in general. Furthermore, the present invention could be used in conjunction with any type of wireless communications device.

It is intended that the invention should be read as being defined by the claims appended hereto, with reference to the description and the attached drawings, but without implying any limitation by such reference.

Claims

CLAIMS: 1. A wireless communications device operable to establish a

wireless communication network with other devices, the wireless communication device comprising medium access control means for controlling access to a communications medium by devices in the network, the medium access control means comprising mode selection means for selecting one of at least two networking modes: in a first of said modes said medium access control means is operable to provide contended access to each device in the network throughout a permitted access period, and in a second of said modes said medium access control means is operable to define two or more groups of devices on the basis of communication activity therebetween, to define two or more sub-periods of said access period, and to allocate each respective group of devices in said network to a corresponding sub-period.

2. A wireless communications device in accordance with claim 1 wherein the medium access control means is operable to determine which of said modes is to be used on the basis of traffic between devices in said network.

3. A wireless communications device in accordance with claim 2 wherein the medium access control means is operable to determine which of said modes is to be used on the basis of at least one of: the uniformity of distribution of traffic levels between devices in said network; the extent to which traffic levels can be determined by said medium access control means; and the overall level of traffic in said network relative to network capacity.

4. A wireless communications device in accordance with claim 3 wherein the medium access control means is operable to select the first mode of operation in the event that: the distribution of traffic levels between devices in said network is substantially uniform; or traffic levels cannot be determined by said medium access control means; or the overall level of traffic in said network is relatively low with respect to network capacity.

5. A wireless communications device in accordance with any preceding claim wherein the medium access control means is operable to select the second mode of operation in the event that traffic in said network exceeds a predetermined level and the traffic in said network can be defined as being predominantly within groups of devices in said network, with relatively low levels of traffic between devices in different groups.

6. A wireless communications device in accordance with any preceding claim wherein the medium access control means is operable in said second mode to define sub-periods in said available access period, each sub-period being allocated for contention by devices in a respective group of devices.

7. A wireless communications device in accordance with any preceding claim wherein the device is operable in accordance with at least one of the IEEE8O2. 11 family of Standards.

8. A method of controlling access by wireless communications devices to a communications medium in a wireless communications network, the method comprising selecting one of at least two networking modes: in a first of said modes said method comprising providing contended access to each device in the network throughout a permitted access period, and in a second of said modes said method comprising defining two or more groups of devices on the basis of communication activity therebetween, defining two or more sub-periods of said access period, and allocating each respective group of devices in said network to a corresponding sub-period.

9. A method in accordance with claim 8 including determining which of said modes is to be used on the basis of traffic between devices in said network.

10. A method in accordance with claim 9 including determining which of said modes is to be used on the basis of at least one of: the uniformity of distribution of traffic levels between devices in said network; the extent to which traffic levels can be determined by said medium access control means; and the overall level of traffic in said network relative to network capacity.

11. A method in accordance with claim 10 wherein the step of selecting one of at least two networking modes is configured such that the first mode of operation is selected in the event that: the distribution of traffic levels between devices in said network is substantiallY uniform; or traffic levels cannot be determined by said medium access control means; or the overall level of traffic in said network is relatively low with respect to network capacity.

12. A method in accordance with any of claims 8 to 11 wherein the step of selecting one of at least two networking modes is configured such that the second mode of operation is selected in the event that traffic in said network exceeds a predetermined level and the traffic in said network can be defined as being predominantly within groups of devices in said network, with relatively low levels of traffic between devices in different groups.

13. A method in accordance with any of claims 8 to 12 wherein in said second mode said method comprises defining sub-periods in said available access period, each sub-period being allocated for contention by devices in a respective group of devices.

14. A method in accordance with any of claims 8 to 13 wherein the method is in accordance with at least one of the IEEE8O2.11 family of Standards.

15. A computer program product comprising processor executable instructions operable to cause a computer to become configured to perform a method in accordance with any one of claims 8 to 14.

16. A wireless communications network comprising a wireless access controller, the wireless access controller being a wireless communications device in accordance with any of claims ito 7.

17. A wireless communications network comprising a plurality of devices, and a wireless access control facility distributed between at least two thereof, the wireless access control facility being operable in accordance with any of claims 8 to 14.

18. A computer program product comprising processor executable instructions operable to cause a computer to become configured to cooperate with likewise configured computers such that said computers are caused to become configured as a wireless communications network in accordance with claim 17.